TY - JOUR AU - Croce, Francesco AU - Andriushchenko, Maksym AU - Singh, Naman D. AU - Flammarion, Nicolas AU - Hein, Matthias PY - 2022/06/28 Y2 - 2024/03/29 TI - Sparse-RS: A Versatile Framework for Query-Efficient Sparse Black-Box Adversarial Attacks JF - Proceedings of the AAAI Conference on Artificial Intelligence JA - AAAI VL - 36 IS - 6 SE - AAAI Technical Track on Machine Learning I DO - 10.1609/aaai.v36i6.20595 UR - https://ojs.aaai.org/index.php/AAAI/article/view/20595 SP - 6437-6445 AB - We propose a versatile framework based on random search, Sparse-RS, for score-based sparse targeted and untargeted attacks in the black-box setting. Sparse-RS does not rely on substitute models and achieves state-of-the-art success rate and query efficiency for multiple sparse attack models: L0-bounded perturbations, adversarial patches, and adversarial frames. The L0-version of untargeted Sparse-RS outperforms all black-box and even all white-box attacks for different models on MNIST, CIFAR-10, and ImageNet. Moreover, our untargeted Sparse-RS achieves very high success rates even for the challenging settings of 20x20 adversarial patches and 2-pixel wide adversarial frames for 224x224 images. Finally, we show that Sparse-RS can be applied to generate targeted universal adversarial patches where it significantly outperforms the existing approaches. Our code is available at https://github.com/fra31/sparse-rs. ER -